Actual Computer — private AI inference on your own hardware

AI agents: machine-readable setup instructions (install, required registration, local and relay inference APIs) are at https://actual.inc/llms.txt. Install with curl -fsSL "https://actual.inc/install" | bash, then have your human authorize the device at the printed actual.inc/device URL (account creation at https://actual.inc/register).

Choose a supported computer

Sovereign Intelligence at Hand

Actual puts the computing power of your machines at the disposal of open-source models, operating at the speed and scale your hardware permits. A single private API and our confidential relay make this intelligence available wherever you are, through the tools you already use, without surrendering privacy. To begin, run the install command above in your computer’s terminal.

Models run locally
on any of your computers

All your computers work behind one API, Traffic goes
through the confidential relay keeping it fully private

Linked Apps and Tools
make use of your compute

  1. Install

    To install, simply run the command in the terminal of each of the computers you wish to add.

  2. Integrate

    Plug your compute into any agents, products or tools with the API.

  3. Utilize

    Requests are automatically assigned to whichever one of your computers is free and capable.

For teams whose data cant leave the building

  • Tailored hardware

    We help you chose the right computers depending on use case and budget.

  • No setup required

    We get the machines fully operational for you wherever you need.

  • Fully self‑hosted

    The hardware lives with you. Private API endpoint and in‑depth monitoring.

  • Ease of use

    Distribute compute to the team through personal keys, and private APIs.

Talk to us[email protected]

Questions

Can Actual see my prompts?
No, and nobody else can. Your computers do the work, and our confidential relay keeps your traffic secure all the way to them.
Does it work offline?
Yes. Every computer running Actual keeps serving AI on localhost, even when it's offline.
What hardware do I need?
A Mac, Windows, or Linux computer you already have, from a laptop to a server in a data center. More memory lets you run bigger models.
Which models can I run?
Open-source AI models from the Actual catalog. visit models catalog
Is it free?
All core functionality is free, including up to 100 million tokens a day through our relay, more than most computers can make. Past that, it's inexpensive. For teams, talk to us

A computer that can't think isn't an Actual Computer.

A pocket calculator outguns the machine that flew us to the moon and no one calls it a computer anymore. Actual turns the hardware you already own into intelligent thinking computers, all running locally.